2006-04-012024-05-15https://scholars.lib.ntu.edu.tw/handle/123456789/663593摘要：台灣地區經常遭受地震、颱風洪水與土流流災害之侵襲，此等災害對建築結構物之損壞 與國人生命財產之影響極大，亟需發展更為先進之健康診斷監測技術與系統，以便對此 等建築結構之損壞與災害進行即時的健康監測、分析評估與早期預警。本研究包括光纖 光柵監測系統之改良、自充電無線監測網路系統之建構與類神經網路，此三項工作可提 供建築結構工程、水利工程和大地工程之結構物和環境進行即時監測、安全評估、分析 與預警功能。此外，本系統所建構之無線監測網路可用於經濟部和交通部所大力推動之 環台灣無線網路監測計畫(M-Taiwan,U-Taiwan)，相關資訊可即時提供政府決策與工程 人員於地震、水災、土石崩塌、水石流災害之現地相關即時資訊，作為防減救災緊急政 策施政之依據。<br> Abstract: Natural disasters exert an enormous toll on development. It is well known that earthquake, scour and debris flow are the major causes for bridge failure in Taiwan. The civil engineering community is becoming increasingly interested in monitoring structural behavior and in assessing its corresponding integration. A bridge health monitoring system, composed of the advanced sensing technology and neural networks, especially suit for harsh environment, is proposed for the unique demand in the project. The new-developed system will be composed of three parts: improved Fiber Bragg Grating sensors (FBG), self-powered wireless sensor networks (WSN) and neural networks. The study will focus on developing a robust structural health monitoring system and environmental early warning system for practical application. The self-powered smart SHM sensing system proposed herein can be used not only for structural engineering but it also benefits to application in the fields of hydraulic engineering for the erosion and variation monitoring of a live-bed during flood, and geologic engineering using as a early warning system for landslide and debris flow. Moreover, WiMax network is the most important project for the Ministry of Economic Affairs (MOEA) and the Ministry of Transportation and Communication (MOTC) to forming the mobile (M-Taiwan) and ubiquitous (U-Taiwan) WSN system. This self-powered smart SHM sensing system proposed herein using Zigbee/WiMax protocol providing real-time information will help engineers and governor for structures maintenance and operation under natural disasters.健康診斷系統自充電無線監測網路類神經網路structural health monitoring systemself-powered wireless sensor networksneural network